META ANALYSIS OF THE RELATIONSHIP BETWEEN CORPORATE SOCIAL RESPONSIBILITY AND TAX AGGRESSIVENESS

Objective: The objective of this study is to investigate the relationship between corporate social responsibility and tax aggressiveness, with the aim of looking at the firmness of the relationship. Theoretical Framework: This study aims to examine the strength and reliability of the connection between corporate social responsibility (CSR) and aggressive tax dodging actions through the utilization of a meta-analysis methodology. Method: The quantitative method used is a meta-analysis approach of 61 studies and 228 observations. Results and Discussion: The results of this study found a firm relationship between corporate social responsibility and tax aggressiveness. Through a meta-analysis, the results of the negative relationship between CSR and tax agressiveness were obtained. The next result was to confirm that the heterogeneity of previous research results related to CSR, and tax agresiveness was due to the diversity of CSR measurement strategies and characteristics of the state legal system. Research Implications: This research has provided input into the fields of accounting and taxation by giving confidence in the relationship between aggressive tax behavior and the existence of disclosures of social responsibility Originality/Value: This research is one of several studies using a meta-analysis approach. Different from other topics raised using this approach, in this research, the emphasis of the study is on the disclosure of CSR, which is correlated with the degree of the tendency of aggressive tax behavior or activities. The study also uses moderators that have not been used in previous meta-analysis studies, namely state law characteristics.


INTRODUCTION
Corporate tax aggressiveness has become increasingly popular as a subject of discussion in recent years (Hanlon & Heitzman, 2010;Dyreng et al., 2016;Wilde & Wilson, 2018;Beer et al., 2020).International media coverage of the very low value of tax payments by a number of multinationals has caused public anger and these various evasion events have made the public aware of the sophisticated structures these companies have created to avoid taxation in large quantities.(Berrong, 2010;Lokanan, 2023;Sandell, 2012).Tax aggressiveness provides a space for companies to enjoy the benefits or facilities of the state without bearing the burden and burden of taxes passed on to other parties namely individuals and consumers.(Amidu et al.,2016).In this study, tax aggressiveness is associated with all tax planning activities, whether legal, illegal, or in the grey territory (Lanis & Richardson, 2012).As for the purpose of tax aggressiveness policy is to reduce corporate taxes (Chen et al., 2010;Frank et al.,2009;Lanis & Richardson, 2012).Previous research has shown that tax aggressiveness can reduce company costs and increase shareholder wealth (Hanlon & Heitzman, 2010).Thus, in order to determine the extent of tax aggressiveness, companies should compare marginal benefits to company marginal costs (Chen et al., 2010).One of the marginal advantages is greater tax savings, whereas marginal expenses include potential fines imposed by tax administrations, implementation costs (time/effort and transaction costs for carrying out tax transactions), and agency fees that are unwilling to engage in tax aggressive activities.(Chen et al., 2010).
The costs and benefits of this tax evasion are linked to the existence of the stakeholders.
While some stakeholders may benefit from tax evasion, others may have to bear the cost.For example, when viewed from the shareholder side, as one of my stakeholders, the benefits obtained from tax evasion are through higher dividends, while for debtholder, one of the other stakeholders, tends to disagree with the tax aggressiveness of a company so that the company can borrow a number of funds to them.(Ayers et al., 2010).The differential consequences on the costs and benefits of this tax evasion make the involvement of all stakeholders, both shareholders, creditors, and the public, relevant.
Research results that are inconsistent and tend to contradict each other are interesting for follow-up.Efforts are also being made to synthesize the results of research, but such research is often limited to a few factors, or based on a particular paradigm.With limited knowledge, there is no comprehensive research, using meta-analyses and maximizing individual research results with various methods of data collection.Meta-analysis studies related to the impact of CSR and tax aggressiveness are aimed at providing confidence in the results of the research (robustness).In addition to obtaining a robust conclusion about the relationship between CSR to tax agresiveness, the study will also look for factors that cause the results between the research, on the relationship of CS R and tax agresivity, to differ.The variable that is supposed to cause the difference in the outcome is called the moderator variable.The commonly used variable to guess the cause of the difference is the measurement of the research variable, the sample is limited to a particular country, and the country's legal system.(Whait et al.,2018;Eny, 2013;Garcia-Meca dan Sanchez-Ballesta, 2009, 2010;Orlitzky dkk., 2003;Souissi dan Khlif, 2012)

TAX AGGRESSIVENESS
There is no universally agreed definition of tax avoidance, tax aggressiveness or tax evasion.Denis Healey, a politician and economist who also served as British Finance Minister from 1974 to 1979, made a statement, "The difference between tax avoidance and tax evasion is just "a thick prison wall."(Elliffe, 2011).Some experts give a definition of the third, one of the definitions that is commonly used is as stated by Slemrod (2004) where it is said that tax avoidance is everything that a company does in order to reduce its tax liability.Another view is expressed by Hanlon & Heitzman (2010) which explains that tax evasion can be seen as a set of tax planning strategies demonstrated by legal strategies and tax fraud.In law, according to Hasseldine & Morris (2013) if viewed from a jurisprudence perspective there is a fundamental difference between tax fraude and avoidance according to which the notion of tax eviction refers to legitimate behavior, whereas the concept of fiscal fraud refers on conduct that violates the code of ethics of taxation.Measurement of tax aggressiveness as revealed by Hanlon (2005) can use several ways, namely by using the Effective Tax Rate (ETR), Book Tax Difference (BTD), Residual Tax Differing (RTC), and Cash Effective tax Rate (CETR).

CORPORATE SOCIAL RESPONSIBILITY
CSR is the activity of a company that in managing its business to get the maximum profit and the minimum possible losses without forgetting to pay attention to social & environmental effects (Lanis & Richardson) (2012).Based on the above definitions, it can be said that CSR is an important factor in supporting the success and survival of a company.(Laguir et al., 2015).The implementation of CSR has become a trend in its practice because implementing CSR is actually beneficial to the company.There are two reasons why companies are starting to take social responsibility.One is for ethical reasons.The other reason is opportunism.Companies may practice CSR to hide the actual action taken to meet the interests of top management (Hemingway & Maclagan, 2004).
There are a large number of studies suggesting different sizes of CSR [Abbott & Monsen, 1979 ;Carroll, 1979;Aupperle et al., 1985].Currently, what should and should not be included to establish a comprehensive and indicative measure of CSR remains a controversial question for academics, researchers, and policymakers.(Turker, 2009).CSR measurements can use several indicators.The indicator used by Orlitzky et al. (2003) is by classifying measurements in four aspects, namely disclosure, reputation rating, social audit, managerial CSR principle and observable & outcame processes.

THE RELATIONSHIP OF CORPORATE SOCIAL RESPONSIBILITY AND TAX
AGGRESIVENESS Public concerns about social and environmental issues have increased over the last few decades; these concerns have also been shifted to companies, where behaviour focused on environmentally friendly and social activities is expected.Over the past few years, the number of companies involved in voluntary CSR strategies has increased rapidly.It drew considerable attention from research communities such as Chen et al., (2017), Hoi et al. (2013) and Lin et al.. (2017).CSR strategy is presented as an alternative management model that emerges and considers companies as a series of relationships, not only between owners and managers but also with parties or groups interested in the evolution of the company: employees, customers, suppliers, competitors, the environment, and society in general.Moreover, previous research, Orlitzky et al., (2003), agreed to maintain the belief that one of the most important consequences of a proactive CSR strategy is improving the company's reputation.This 6 stakeholders about the company.In this case, in addition to social and environmental issues, there is also concern about the high percentage of taxes that will be considered as a cost by the company which will affect the reduction of shareholder wealth.The growth of literature on CSR and tax aggressiveness has recently been considerable, however, empirically empirical academic researchers have not yet found a consensus in their relationship (Laguir et al., 2015;Landry et al. 2013;Lin and al., 2017).
On the one hand, previous literature suggests that CSR disclosure and tax aggressiveness can also be harmonized.Although the primary objective of the company is to maximize shareholder value, Hoi et al. (2013) and Laguir and al. (2015) confirm that companies should strive to meet all stakeholder demands to improve their image and reputation.(Hill & Jones, 1992;Minnick & Noga, 2010).On the other hand, literature states that corporate taxes can only be attributed to social and environmental performance when their payments imply benefits to society.In this regard, Landry et al. (2013) affirmed that tax savings resulting from aggressive strategy planning can be far cheaper than the cost of loss of reputation that may be incurred by tax austerity activities.CSR can increase shareholder value because the negative effects of tax evasion on shareholders can be offset by social and environmental activities (Hoi et al., 2013;Laguir et al. 2015), but they still prefer to support tax savings and thus increase their wealth.Thus, the negative link between CSR and tax evasion is supported by this approach.The greater the implementation of CSR, those, the lower corporate tax evasion.Based on the above analysis, this study proposes a further hypothesis: H1.The more socially responsible a company will reduce the level of tax aggressiveness

POTENTIAL MODERATORS
One of the purposes of this research is to identify the things that cause heterogeneity through a moderator variable analysis.The variable that the moderator allegedly affects the relationship between CSR and tax aggressiveness is associated with the measurement of variables and samples in this case related to the application of the state legal system (Whait, 2008).

Measurement of Corporate Social Responsibility
In line with previous empirical research on the issue of CSR disclosure (Orlitzky et al., 2003;Swanson and Orlitzki, 2017;Visser et al. 2010) 7 revelation measures as independent variables (e.g.social audits of corporate CSR strategies), CSR ratings given by external agencies such as Kinder, Lydenberg, and Domini (KLD), observable results emerging from them, social and environmental accounting practices, entered companies in socially responsible indices.More importantly, these CSR measures do not capture the multidimensional nature of this theoretical construction (Huseynov and Klamm, 2012;Laguir et al., 2015), which covers social, environmental, and economic issues.According to Freedman (2003), one of the methods or approaches that can be taken to CSR disclosure is by making social responsibility disclosures in Annual Reports.The measuring instrument in CSR disclosure that can be used is the Corporate Social Responsibility Index (CSRI).Unlike Freedman et al. (1997) for instance, it uses two perception-based CSR measurements, the KLD and Fortune reputation surveys, and two fact-based measures, the GRI and philanthropy.As such, the current study proposes the following hypothesis: H2.Measurement of corporate social responsibility moderates the relationship between CSR and tax aggressiveness.

Characteristics of The State Legal System
The existence of state law affects CSR in every company (Salhi et al., 2019) The characteristics of the state are reflected in the legal system of each state, namely whether it is common law or civil law.Countries that have similar legal systems tend to have similar CSR processes, thus having the same CSR characteristic tendency.(Liang & Renneboog, 2017).The nature of this multidimensional and externally driven CSR element suggests that CSR must be fundamentally related not only to the company's own choices but also to regulations, institutional arrangements, and public preferences.Moreover, in addition to viewing CSR as a mechanism to deal with externality impacts, CSR is also seen as a more fundamental tradeoff between the focus of shareholders and other stakeholders (at the corporate level) as it is between the rules and the wisdom of the institutions that regulate economic life.(Ferrell et al., 2016).
Such tradeoffs are heavily dependent on corporate contracts with the surrounding environment that are likely to be formed by different legal rules and law enforcement mechanisms in different countries.It underpins legal diversity as an aspect that leads to a difference in outcomes in the relationship between CSR and tax aggressiveness, because tax inconsistencies also relate to law enforcement in a country.Previous research has focused on a sample of companies that are limited to specific countries, such as France (Laguir et al., 2015), the United States (Lanis andRichardson, 2011, 2015;LANIS et al., 2017;Watson, 2015), Australia (Lanis and Richardson and al., 2013;Taylor et al, 2015), Canada (Landry et al.;2013;Zeng, 2016) or Indonesia.(Simorangkir et al, 2012).As such, the current study proposes the following hypothesis: H3. Country characteristics based on the legal system moderate the relationship between CSR disclosure and tax aggressiveness

LITERATURE SEARCH
The literature search was carried out using various terminologies but different definitions of CSR and tax aggressiveness collected from various articles.The following keywords are used to search for terms for CSR: "corporate social responsibility," "corporate social reporting," "sustainability disclosure," and "sustainability reporting."Keywords for tax aggressiveness include: "tax aggressiveness", "tax avoidance", "tax sheltering", "tax evasion", "tax management" and "tax risk management" which come from various articles.Electronic searches of Scopus, ProQuest, EBSCOHost, Emerald Insight, Science Direct, JSTOR, SpringerLink, Taylor & Francis Online, Wiley Online Library, and Sinta Indonesia in the period 2010 -2022.

FILTER CRITERIA
The study was chosen to be included in the meta-analysis based on the following criteria: (1) an empirical study that reports the correlation coefficient between CSR and tax aggressiveness, (2) following the recommendations of Peterson & Brown, (2005) that the maximum number of "effect sizes" that may have to be included to increase the generalization of meta-analysis findings, we include all empirical studies that report relevant statistics (t values, F-ratios, Chi-squared, mean and SD, standardized regression coefficients (beta) that explicitly or indirectly estimate the relationship between CSR disclosure and tax aggressiveness.The author then transforms indirect statistical data (other than r) to r using the formula according to (Rosenthal, 1991).The articles used as data in this research are those found in Emerald, Science Direct, Proquest and EBSCO search engines.After completing this procedure, a total of 61 papers that fulfill these criteria were acquired.The research included the flowchart illustrating the method described above.The list of articles used in the meta analysis process is attached in the appendix.

Diagram of the inclusion and exclusion of research.
Source: Author (2024)

CODING PROSEDUR
Every study collected will get coding using the author-centric postulated approach (Webster & Watson, 2002).To avoid coding errors, the authors create a coding stream that determines the information to be filtered from each study as a reference,sample size, the country where the study was conducted, journal name, a correlation between research variables.

THE STAGES OF DATA PROCESSING
Meta-analysis is one method of research that combines several research results with the same material, but various conclusions.Meta-analysis can be used as a method or tool for synthesizing or combining quantitative research results.Further meta-analysis is used to confirm the variation of findings substantively or methodologically to describe differences between studies (Cooper, 2017).Referring to (Cooper, 2017) which states that there are four stages carried out in the meta-analysis approach.These stages are the steps that must be taken in the meta-analysis include the following procedures (1) Identify and formulate research problems (2) Gathering data through a selection of articles or research results that are relevant to the research problem (3) Explanation and evaluation of data and (4) Analysis and interpretation of the results of the analysis itself.According to Cooper (2017), after obtaining the r statistics for each study, there are three further steps to estimate the correlation and average variance.(1) Find the average correlation coefficient (r).( 2) Compute the variance of all individual correlation coefficients using a weighted average error squaring the sample size.(3) Calculating the estimated variation in sampling error.The 95% confidence level is used to determine the significance level of the mean correlation coefficient.The next step is to examine heterogeneity through I 2 values (Higgins et al., 2003).The higher the value of I 2 , the more heterogeneity it shows.

HETEROGENEITY TEST AND PUBLICATION BIAS ANALYSIS
The Q value, I 2 and Ƭ² were used to judge the heterogeneity of coded studies.The result shows that Q = 4158.325[df(Q) = 227, p < 0.05], which is greater than the chi-square critical value (df(Q) = 227) at 95% confidence interval level, indicating that there was a high degree of heterogeneity in primary researches (Higgins et al., 2003).The I 2 = 94.54%> 75%, indicating that 94,54% of the observed variation was caused by the difference of the true value (Lipsey and Wilson, 2008).The Ƭ² result further confirms the accuracy of the random-effect model that was chosen.The Ƭ² = 0.0064, where this value is in the 95% confidence interval range, namely 0.00628 -0.00682 and the heterogeneity test was of statistical significance (Higgins et al., 2003).Publication bias is the potential occurrence where not all research conducted on a particular issue are made available to the public, and these unpublished studies may differ in a systematic way from the ones that are published.The fail-safe N is defined as the minimum number of unpublished studies needed to reduce the cumulative effect of meta-analysis to a nonsignificant level (Rosenthal, 1979).The fail-safe N of the current study is 1150, which is far greater than the critical value for N, 5k þ 10 (k = 228).So, it is less likely to reverse the conclusion of the meta-analysis.

ANALYSIS RESULT
Table 1 displays the primary effect size, which was 0,0193.Studies show that the "CSR" indicator has a positive and significant correlation with the measurement indicator of tax aggressiveness.The positive association between CSR and this tax aggressiveness 11 measurement indicator has the opposite meaning to tax aggressivity.So through this study it can be established that there is a low negative correlation between CSR and Tax Aggressiveness (TA).The 95% confidence intervals (0.0074 and 0.0312) excluded zero and the p-value was less than 0.001, indicating a significant result.Therefore, there is sufficient evidence to support the first hypothesis (H1).The next stage is to test whether heterogeneity exists to provide justification for testing other hypotheses.The determination of the occurrence of heterogeneity can be done using the parameters Q and p-value and reinforced by Ƭ² and I 2 .The test results showed that the overall indicator supports the occurence of heterogenity is the value of Q of 4158,325 with p-value < 0,000, the value of Ƭ² is 0,0064, with the confidence interval at the level of 95% significance is 0.00628-0,00682, and the value I 2 of 94,54% with the interval of confidence 94,09% -94,55%.The findings of the moderator analysis are presented in Table 2.The test results for hypothesis 2 showed the moderator effect of the diversity of CSR measurements.After mapping, CSR measurements are grouped into three groups as shown in table 2. Based on the results of testing using the 3 subs-groups, as shown in table 2, there was only one group, disclosure measurement strategy (DIS), that produced a positive correlation value.The value of the positive coefficient in DIS, means that there is a negative relationship between CRS and tax aggressiveness with a correlation value of 0.06621 at the minimum confidence interval of 0.03783 and a maximum of 0.09478.Test results for two other measurement strategies, namely reputation (REP) and social audit (SA), showed insignificant results.The test findings for hypothesis 3 indicated the moderating influence of the diversity of characteristics of the state legal system.After mapping, characteristics of the state legal system are classified into two groups as shown in table 3. The results of each group indicated diverse results.Based on the results of testing employing the 2 subs-groups, as indicated in table3, there was only one group, Civil Law (CIL), that provided a positive correlation value.The value of the positive coefficient in CIL, indicating that there is a negative association between CRS and tax aggressiveness with a correlation value of 0.06235 at the minimum confidence interval of 0.01866 and a maximum of 0.08613.Test results for the other one, common law (COL), showed insignificant result.

DISCUSION
The results show that it can be firmly proven that CSR correlates negatively with tax aggressiveness.The negative relationship between CSR and tax aggressiveness as a whole also shows that CSR practices are appreciated by stakeholders and can improve the company's reputation.Corporate involvement in social responsibility that gives this positive image does not want to be destroyed by acts of tax aggressiveness.Thus, the results of Meta Analysis For moderator analysis, Meta Analysis results show that the heterogeneity of the CSR relationship and tax aggressiveness is also due to highly varied CSR measurements.Meta analysis revealed the tendency of previous researchers to measure CSR based on four CSR measurement strategies, namely content analysis, reputation, and social audit.Strategies for measuring with reputation tend to use the results of third-party assessments such as independent CSR assessment agencies or standard assessment standards, while measurements with disclosure are more adaptive in practice but still follow generally applicable disclosures, such as GRI.Sub-group analysis shows that with adaptive measurements the relationship between CSR and tax aggressiveness is positive, but this does not apply to the measurement with reputation and social audit, through both of these measures obtained CSR results are not related to tax agresiveness.The results of this analysis indicate the need for a more adaptive measured in revealing CSR relationships and tax agresivity.CSR relationship and tax agressiveness less can be revealed with less flexible CSR measures.
Sub-group analysis based on state characteristics showed different results in civil and common law states.Empirical evidence in civil law countries showed negative links between CSR and tax aggressiveness, whereas in common law countries it was not proven that CSR was linked to tax agressiveness.The existence of state law affects CSR in every company (Salhi et al., 2019) The characteristics of the state are reflected in the legal system of each state, namely whether it is common law or civil law.Countries that have similar legal systems tend to have similar CSR processes, thus having the same CSR characteristic tendency.(Liang & Renneboog, 2017).The nature of this multidimensional and externally driven CSR element suggests that CSR must be fundamentally related not only to the company's own choices but also to regulations, institutional settings, and public preferences.
Moreover, in addition to viewing CSR as a mechanism to deal with externality impacts, CSR is also seen as a more fundamental tradeoff between the focus of shareholders and other stakeholders (at the corporate level) as it is between the rules and the wisdom of the institutions that regulate economic life.(Ferrell et al., 2016).Such tradeoffs are heavily dependent on corporate contracts with the surrounding environment that are likely to be formed by different 14 of legal practice as an aspect that leads to differences in outcomes in the relationship between CSR and tax aggressiveness, as tax inconsistencies are also related to the law-enforcement in a country.

CONCLUSION
The study aims to test the relationship between CSR and tax aggressiveness through meta-analysis using a set of data from various previous studies.There are 61 articles that produce 228 research contexts outlined from previous research.Based on the evidence obtained, it is concluded that in all contexts CSR reduces tax aggressiveness.This conclusion has a wide generalization power because it is done on various data settings so capturing various research contexts.Further analysis shows that the relationship between CSR and tax aggressiveness is highly varied or highly heterogeneous.The heterogeneity of the relationship between CSR and tax aggressiveness reveals the existence of moderation that can affect the results of the CSR relationship test and tax agresiveness, i.e. the versatility of CSR measures, and the variability of characteristics of the state legal system.
In measuring CSR, disclosure measurement strategies result in a significant positive relationship between CSR and tax aggressiveness while reputation strategies and social audit strategies do not result in any significant link between CSR and tax agresiveness.The characteristics of the state legal system are a moderator of the relationship between CSR and tax aggressiveness.Meta-analysis results show that in civil law countries, CSR increases the effective tax rate or, in other words, minimizes tax agresiveness, but it is different in common law countries where CSR does not increase tax aggressivity.These findings prove that characteristics of the state legal system can influence the significance and direction of the relationship between CSR and tax aggressiveness.
reputation/image leads to a positive reinforcement of the perceptions and beliefs of my Meta Analysis of The Relationship Between Corporate Social Responsibility and Tax Aggressiveness ___________________________________________________________________________ Rev. Gest.Soc.Ambient.| Miami | v.18.n.5 | p.1-21 | e06699 | 2024.
strategy; DIS = disclosure; REP = reputation; SA = social audit; k = observation; ρ = correlation; *p-value <0.05 Source: Author (2024) ___________________________________________________________________________ Rev. Gest.Soc.Ambient.| Miami | v.18.n.5 | p.1-21 | e06699 | 2024.13 confirm the work of the Instrument Stakeholder Theory and Legitimation Theory in CSR practices.Although generally, CSR and tax aggressiveness are linked negatively, analysis suggests that the relationship is far from homogeneous or heterogenous.Based on the results of previously presented tests, the factors that can lead to the occurrence of heterogeneity are the CSR measurements and characteristics of the state legal system (Civil Law or Common Law).
legal rules and law enforcement mechanisms in different countries.This underpins the diversity Meta Analysis of The Relationship Between Corporate Social Responsibility and Tax Aggressiveness ___________________________________________________________________________ Rev. Gest.Soc.Ambient.| Miami | v.18.n.5 | p.1-21 | e06699 | 2024.

Table 1
Analyst Impact of CSR and Tax Aggressiveness

Table 2
Analyst Results Sub-group CSR Measurement Strategy

Table 3
Analyst Results Sub-group Characteristics of the state legal system